Studies on Pyridines: I. The Basicity of Pyridine Bases

hyperconjugative effect alone should make both 2- and4-picoline stronger bases than pyridine, the +1 effect should be greater in the 2-position, due t...
1 downloads 0 Views 296KB Size
[CONTRIBUTION

FROM THE

DEPARTMENT O F CHEMISTRY, VILLANOVA

COLLEGE]

STUDIES ON PYRIDIKES: I. T H E BASICITY O F PYRIDIKE BASES ALEXANDER CERO

AND

JAMES J. MARKHAiLIl

Received June 26, 1961

We were interested in studying the base strength of pyridine and its homologs, particularly those methylpyridines which contain the methyl group attached to the 2- and 4-positions. These positions are made electron-deficient by the resonance of the pyridine ring

w

and it mas thought worthwhile to investigate what effect an electron-donor methyl group mould exercise in those positions. Brown and Barbaras (1) attacked this problem by a study of the addition compounds of pyridine bases with trimethylborine. They point out that, while the hyperconjugative effect alone should make both 2- and 4-picoline stronger bases than pyridine, the +I effect should be greater in the %position, due to its proximity to the nitrogen. The base strengths of these three bases would therefore be expected to increase in the order pyridine, 4-picoline, 2-picoline. Actually, while pyridine and 4-picoline behave as expected, the addition compound of trimethylborine with 2-picoline dissociates much more readily than the corresponding adducts with the other bases, and Brown and Barbaras attribute this to steric hindrance due to the bulk of the methyl group which counteracts its electrondonor effect when in the 2-position. ?Tie attempted to determine the aqueous ionization constants of pyridine and some of its homologs from their pH titration curves. Results of direct determinations of these ionization constants can be found in the literature ( 2 ) but agreement between various authors is not good. This is only to be expected since pure pyridine bases were not easily available and since the carbon dioxide error is liable t o interfere greatiy with measurements on such weak bases. In order to overcome these difficulties, we purified our bases carefully and devised an experimental technique aimed a t the complete exclusion of atmospheric carbon dioxide. As a further measure of safet8ywe disregarded the pH of the aqueous solutions of the bases themselves (where error due to COzmay be great) as well as the pH at the inflection point of the titration curve (where considerable error is likewise possible due to the steep drop of the pH) and took our results from the point of half-neutralization where [B] = [BH+] and therefore the pH of the solution is equal to the p& of the conjugate acid of the base titrated. These p K a values and the corresponding &'s are shown in Table I. 1 Present address : Department of Oceanography, Agricultural & bIechsnica1 College of Texas, College Station, Texaa. 1535

1836

ALEXANDER GERO AND JAMES J. MARKNAM

An interesting regularity becomes apparent if we plot the above p K A values against the number of methyl groups in the molecule. As seen in Fig. 1 the points

Pyridine. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-Picoline. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-Picoline.. . . . . 2,6-Lutidine.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2,4-Lutidine.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2,4,6-Collidine.. . . . . . . . ......

1.7 x 9.1 x 1.1 x 4.2 X 6.1 x 2.8 x

5.23 6.05 6.62 6.79 7.45

-.I ~

10-9 10-9 10-8

10-8 10-8 10-7

7a

6.5

Pb 6.0

5.5

5.0

,

I

1

t

0

I

2

3

NUMBER OF METHYL GROUPS

FIGURE 1. P L O T

OF

PICA

?JS

NCMBER

OF & h H Y L

GROWS O F

PYRIDINE

BASES

obtained lie very close to a straight line which means, since this is a logarithmic diagram, that each methyl group multiplies the fundamental basicity of pyridine by the same factor (about 6). Closer inspection reveals another fact; the points corresponding to bases with 2-methyl groups deviate from the straight line

STUDIES OK PYRIDINES. I

1837

by lying slightly below it while those corresponding to bases with a $-methyl group lie slightly above the straight line. This means that the multiplication factor for 2-methyl groups is somewhat different from that for 4-methyl groups. A mathematical relationship was sought which would express these observations, and it was found that the figures of Table I are solutions of the following equation:

in which pKp, represents the p K A of pyridine, y the number of methyl groups in the y-position and a the number of methyl groups in the a-position. The theoretical interpretation of the equation is clear: a methyl group manifests its electron-donor character by increasing the electron density a t the N atom (expressed in the ~ K Aby ) a constant of 0.73 if the methyl group is in the a-position and 0.82 if it is in the y-position. The fact that the y-constant is greater than the a-constant is reasonably interpreted to demonstrate qualitatively the same steric effect as found by Brown and Barbaras while quantitatively the effect we found is much smaller because the hydronium ion is smaller than the trimethylborine molecule. The presence of two methyl groups, one in the a- and one in the y-position, increases the p G of pyridine (5.23) by both the a- and the y-constants. But if both methyl groups are in the a-position, they increase the fundamental basicity of the ring by less than twice the a-constant, showing that their steric hindrance is more than twice that of only one methyl group and necessitating the correction of 0.03 for the a-constant. This is quite plausible; steric hindrance by one a-methyl group seals off only one side of the nitrogen while two methyl groups in the two a-positions restrict hydronium ions to a narrow avenue of access. It might be objected that the restriction suffered by the hydronium ion may be due to electrostatic repulsion by the positive methyl group as much as to genuine steric hindrance. But no steric hindrance is completely free of electrostatic effects since the outside of any atom or group is an electron cloud. It is therefore reasonable to compare the hydronium ion with the trimethylborine molecule in this context although the former is charged and the latter is not. EXPERIMENTAL 1 . Materials. In the purification of pyridine and the picolines we followed closely the method of Brown and Barbaras (1) The lutidines were generously supplied by the Koppers Company in 95yo purity. For further purification we took advantage of the low solubility of the complexes formed by the lutidines with mercuric chloride (3). The crude lutidines were dissolved i n dilute hydrochloric acid and the solutions slowly stirred into a hot 10% solution of HgC12 in the proportion 3 HgClz t o 1 lutidine. The precipitates were filtered and recrystallized repeatedly from very dilute aqueous HC1 until the melting points were constant (127' for the adduct of 2,4-lutidine, 162" for t h a t of 2,6-lutidine, both uncorrected). Then the bases were liberated with 40% sodium hydroxide, dried with CaO, and distilled. The boiling points were in agreement with earlier d a t a (3). ?3,+$,B-Collidine mas synthesized by the method of Hantzsch (4). After distillation, all bases were stored in small wash bottles over potassium hydroxide I

1838

ALEXANDER GERO AXD JBXEG 3 . YARI